On the Sensitivity of Probabilistic Networks to Reliability Characteristics (original) (raw)
Related papers
Detailing Test Characteristics for Probabilistic Networks
Lecture Notes in Computer Science, 2003
In the medical domain, establishing a diagnosis typically amounts to reasoning about the unobservable truth, based upon a set of indirect observations from diagnostic tests. A diagnostic test may not be perfectly reliable, however. To avoid misdiagnosis, therefore, the reliability characteristics of the test should be taken into account upon reasoning. In this paper, we address the issue of modelling such characteristics in a probabilistic network. We argue that the standard reliability characteristics that are generally available from the literature have to be further detailed, for example by experts, before they can be included in a network. We illustrate this and related modelling issues by means of a real-life probabilistic network in oncology.
Lecture Notes in Computer Science, 2005
This paper addresses issues in constructing a Bayesian Network (BN) domain model for diagnostic purposes from expert knowledge. The novelty of this paper is the approach for structured generation of a model that incorporates the unstructured multifaceted and possibly conflicting probabilistic information provided by the experts. 1 Bayesian or Causal Networks are directed acyclic graphs whose edges indicate causality and whose structure captures the joint probability functions of the node variables.
Why is diagnosis using belief networks insensitive to imprecision in probabilities
1996
Recent research has found that diagnostic performance with Bayesian belief networks is often surprisingly insensitive to imprecision in the numerical probabilities. For example, the authors have recently completed an extensive study in which they applied random noise to the numerical probabilities in a set of belief networks for medical diagnosis, subsets of the CPCS network, a subset of the QMR (Quick Medical Reference) focused on liver and bile diseases. The diagnostic performance in terms of the average probabilities assigned to the actual diseases showed small sensitivity even to large amounts of noise. In this paper, we summarize the findings of this study and discuss possible explanations of this low sensitivity. One reason is that the criterion for performance is average probability of the true hypotheses, rather than average error in probability, which is insensitive to symmetric noise distributions. But, we show that even asymmetric, logodds-normal noise has modest effects. A second reason is that the gold-standard posterior probabilities are often near zero or one, and are little disturbed by noise.
Discrimination and its sensitivity in probabilistic networks
A probabilistic network built for an application domain often has a single output variable of interest, for which either the posterior probability of one of its values or its most likely value is reported and used for subsequent decision making. For our domain of application, however, we are interested primarily in how well the network distinguishes between various compound output values of interest for different diagnostic variables. To capture this, we introduce a concept of discrimination, and illustrate a measure to this end, based upon joint posterior probabilities. In addition, we address the sensitivity of discrimination to inaccuracies in a network's parameters and show that standard sensitivity functions suffice for studying the effects of such inaccuracies.
Medical Diagnosis Using a Probabilistic Causal Network
Applied Artificial Intelligence, 1989
This paper relates our experience in developing a mechanism for reasoning about the di erential diagnosis of cases involving the symptoms of heart failure using a causal model of the cardiovascular hemodynamics with probabilities relating cause to e ect. Since the problem requires the determination of causal mechanism as well as primary cause, the model has many intermediate nodes as well as causal circularities requiring a heuristic approach to evaluating probabilities. The method we have developed builds hypotheses incrementally by adding the highest probability path to each nding to the hypothesis. With a number of enhancements and computational tactics, this method has proven e ective for generating good hypotheses for typical cases in less than a minute.
Impact of Bayesian Network Model Structure on the Accuracy of Medical Diagnostic Systems
Lecture Notes in Computer Science, 2014
While Bayesian network models may contain a handful of numerical parameters that are important for their quality, several empirical studies have confirmed that overall precision of their probabilities is not crucial. In this paper, we study the impact of the structure of a Bayesian network on the precision of medical diagnostic systems. We show that also the structure is not that important-diagnostic accuracy of several medical diagnostic models changes minimally when we subject their structures to such transformations as arc removal and arc reversal.
Dynamic Network Updating Techniques for Diagnostic Reasoning
Uncertainty Proceedings 1991, 1991
A new probabilistic network construction system, DYNASTY, is proposed for diagnos tic reasoning given variables whose probabil ities change over time. Diagnostic reason ing is formulated as a sequential stochastic process, and is modeled using influence dia grams. Given a set 0 of observations, DY NASTY creates an infl uence diagram in or der to devise the best action given 0. Sensi tivity analyses are conducted to determine if the best network has been created, given the uncertainty in network parameters and topol ogy. DYNASTY uses an equivalence class ap proach to provide decision thresholds for the sensitivity analysis. This equivalence-class approach to diagnostic reasoning differenti ates diagnoses only if the required actions are diff erent. A set of network-topology updat ing algorithms are proposed for dynamically updating the network when necessary.
Comparison of Rule-Based and Bayesian Network Approaches in Medical Diagnostic Systems
Lecture Notes in Computer Science, 2001
Almost two decades after the introduction of probabilistic expert systems, their theoretical status, practical use, and experiences are matching those of rule-based expert systems. Since both types of systems are in wide use, it is more than ever important to understand their advantages and drawbacks. We describe a study in which we compare rule-based systems to systems based on Bayesian networks. We present two expert systems for diagnosis of liver disorders that served as the inspiration and vehicle of our study and discuss problems related to knowledge engineering using the two approaches. We finally present the results of a simple experiment comparing the diagnostic performance of each of the systems on a subset of their domain.
Bayesian networks in reliability
Reliability Engineering & System Safety, 2007
Over the last decade, Bayesian Networks (BNs) have become a popular tool for modelling many kinds of statistical problems. We have also seen a growing interest for using BNs in the reliability analysis community. In this paper we will discuss the properties of the modelling framework that make BNs particularly well suited for reliability applications, and point to ongoing research that is relevant for practitioners in reliability.
2008
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e 8 9 ( 2 0 0 8 ) 189-201 a b s t r a c t Causal probabilistic networks provide a natural framework for representation of medical knowledge, allowing clinical experts to encode assumptions about causal dependencies between stochastic variables. Application in medical decision support has produced promising results. However, model features and parameters may vary geo-or demographically.